Sprint 2 retrospective

Jump to bottom

lyashenkogs edited this page Feb 12, 2017 · 2 revisions

The finish of the Sprint 2

Performance improvements:

decrease logging verbosity for big arrays
add optimizations to cutAllStartPostions, to avoid iteration over all array elements by ArrayList.removeAll() method
implement returning of start positions without available steps to a pizza
added a custom utility class for simple performance measures

Summarizing:

Slicing of: example, small, medium slices took 10m 39 seconds on a MacBookPro Results:

example 6
small 30
medium 33037
big - too long to slicing. Was ran and took about 20 hours.(a desktop PC with an Intel-I54670K)

What went well: //TODO

What can be better: //TODO

This algoritm shoud be optimized in future.